JP2020061404A - Substrate processing method and substrate processing apparatus - Google Patents

Abstract

To provide a substrate processing method capable of imaging process liquid in liquid column shape discharged to the end part of a substrate.SOLUTION: A substrate processing method includes a holding step, a rotation step, a lifting step, a beveling step, an imaging step and a monitoring step. In the holding step, a substrate is held at a substrate holding part. In the rotation step, the substrate is rotated by rotating the substrate holding part. In the lifting step, a cup member surrounding the outer boundary of the substrate holding part is lifted, and the upper end of the cup member is located at an upper end position higher than the top face of the substrate held by the substrate holding part. In the beveling step, process liquid is discharged to the end part of the top face of the substrate held by the substrate holding part, from the discharge opening of a nozzle at a position lower than the upper end position. In the imaging step, an imaging region including the process liquid discharged from the discharge opening of the nozzle, viewed from the imaging position above the substrate, is captured by means of a camera, thus obtaining a captured image. In the monitoring step, discharge condition of the process liquid is determined based on the captured image.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a substrate processing method and a substrate processing apparatus.

  As an apparatus for performing processing on a substrate, there is used a substrate processing apparatus that rotates a substrate in a horizontal plane and discharges a processing liquid from a discharge nozzle onto the surface of the substrate. The processing liquid that has arrived at the approximate center of the substrate from the discharge nozzle spreads over the entire surface due to the centrifugal force that accompanies the rotation of the substrate, and then scatters outward from its peripheral edge. By spreading the processing liquid over the entire surface of the substrate, the processing liquid can act on the entire surface of the substrate. As the processing liquid, a chemical liquid or a cleaning liquid suitable for the processing on the substrate is adopted.

  In such a substrate processing apparatus, there is proposed a technique of providing a camera in order to monitor whether or not the processing liquid is appropriately ejected (Patent Documents 1 to 5).

  Further, in the manufacturing process of the semiconductor substrate, various films remaining on the peripheral edge of the substrate may adversely affect the device surface of the substrate.

  Therefore, conventionally, a bevel process for removing the film from the peripheral edge of the substrate has been proposed. In the bevel processing, the substrate is rotated in a horizontal plane, and a processing liquid for removal is discharged from a discharge nozzle to the end portion of the substrate, so that the film at the peripheral edge portion of the substrate is removed by the processing liquid.

JP, 2005-173148, A JP, 2017-29883, A JP, 2015-18848, A JP, 2016-122681, A JP, 2008-135679, A

  In the bevel processing, since the processing liquid only needs to be supplied to the end portion of the substrate, the flow rate of the processing liquid is reduced. That is, the liquid columnar processing liquid discharged from the discharge nozzle becomes thin. Therefore, the liquid columnar processing liquid is easily affected by various factors such as the air flow accompanying the rotation of the substrate and static electricity generated in the surroundings, and the discharge state thereof is likely to change.

  However, in the bevel processing, since the distance between the discharge nozzle and the substrate is narrow, it is necessary to devise a method for imaging the liquid columnar processing liquid discharged from the discharge nozzle.

  Therefore, it is an object of the present application to provide a substrate processing method and a substrate processing apparatus capable of monitoring the liquid columnar processing liquid discharged to the end portion of the substrate.

  A first aspect of the substrate processing method is a holding step of holding a substrate on a substrate holding section, a rotating step of rotating the substrate holding section to rotate the substrate, and a cup member surrounding the outer periphery of the substrate holding section. And the upper end of the cup member is positioned at an upper end position higher than the upper surface of the substrate held by the substrate holding part, and a nozzle outlet located at a position lower than the upper end position. A bevel processing step of discharging a processing liquid to an end portion of an upper surface of the substrate held by the substrate holding part, and an imaging region including the processing liquid discharged from the discharge port of the nozzle, An imaging step in which a camera captures an image of an imaging region viewed from an imaging position above the substrate held by a holding unit, and an ejection state of the processing liquid is determined based on the imaging image. Monitoring process Equipped with a.

  A second aspect of the substrate processing method is the substrate processing method according to the first aspect, wherein in the monitoring step, the ejection is positioned immediately below the nozzle in the captured image and is longer in the horizontal direction than in the vertical direction. It is determined whether or not the first statistic of the brightness value of the pixel in the determination region is equal to or more than a first threshold value, and when the first statistic amount is equal to or more than the first threshold value, the treatment liquid is discharged from the nozzle. It includes a step of determining that the ink is being discharged.

  A third aspect of the substrate processing method is the substrate processing method according to the second aspect, wherein the monitoring step determines whether or not the first statistic is equal to or greater than a second threshold value that is greater than the first threshold value. And the step of determining that the treatment liquid is splashing on the upper surface of the substrate when the first statistic is equal to or more than the second threshold value.

  A fourth aspect of the substrate processing method is the substrate processing method according to the second or third aspect, wherein in the monitoring step, a rotation direction of the substrate with respect to the ejection determination region in the captured image is detected. The method includes a step of determining whether or not the treatment liquid is splashed by the liquid splashing on the upper surface of the substrate based on the brightness value of the pixel in the liquid splash determination region set on the downstream side.

  A fifth aspect of the substrate processing method is the substrate processing method according to the fourth aspect, wherein in the captured image, the liquid is provided on the upstream side in the rotation direction of the substrate with respect to the ejection determination region. The judgment area is not set.

  A sixth aspect of the substrate processing method is the substrate processing method according to any one of the second to fifth aspects, wherein the ejection determination region is reflected on the upper surface of the substrate by specular reflection in the captured image. The position is set to include a part of the processing liquid.

  A seventh aspect of the substrate processing method is the substrate processing method according to the sixth aspect, wherein an exposure time of the camera is set to a time required for one rotation of the substrate or more.

  An eighth aspect of the substrate processing method is the substrate processing method according to the sixth aspect, wherein a plurality of picked-up images acquired by the camera are integrated within a time period required for one rotation of the substrate or more. Alternatively, the first statistic is calculated for the brightness values of the pixels in the ejection determination area in the captured images obtained by averaging.

  A ninth aspect of the substrate processing method is the substrate processing method according to the first aspect, wherein in the monitoring step, the captured image is discharged from the tip of the nozzle by a classifier that has undergone machine learning. The liquid is classified into categories that indicate the discharge state.

  A tenth aspect of the substrate processing method is the substrate processing method according to the ninth aspect, wherein in the monitoring step, ejection that is positioned immediately below the nozzle and that is longer in the horizontal direction than in the vertical direction in the captured image is performed. A judgment area is cut out, and the cut-out image of the discharge judgment area is input to the classifier.

  An eleventh aspect of the substrate processing method is the substrate processing method according to any one of the first to tenth aspects, wherein the captured image includes a part of a peripheral edge of the substrate, The bevel processing step is a step of obtaining a substrate peripheral edge position of a part of the peripheral edge of the substrate based on the captured image, and moving the nozzle from the substrate peripheral edge position to a processing position in a central portion of the substrate by a predetermined width. Including steps.

  An aspect of the substrate processing apparatus is to hold a substrate and rotate the substrate, a substrate holding portion, a cup member that surrounds an outer periphery of the substrate holding portion, and raise the cup member so that the upper end of the cup member is The substrate holding unit has an elevating mechanism which is located at an upper end position higher than the upper surface of the substrate, and an ejection port which is located at a position lower than the upper end position, and is held by the substrate holding unit from the ejection port. An imaging area including a nozzle for ejecting a treatment liquid at an end portion of the upper surface of the substrate and a treatment liquid ejected from the ejection port of the nozzle, which is above the substrate held by the substrate holder. A camera that captures a captured image by capturing an imaged region viewed from the image capturing position, and an image processing unit that determines the ejection state of the treatment liquid based on the captured image.

  According to the first aspect of the substrate processing method and the aspect of the substrate processing apparatus, the processing liquid ejected from the nozzle can be imaged, and the ejection state can be appropriately determined based on the imaged image.

  According to the second aspect of the substrate processing method, it is possible to determine whether or not the processing liquid is discharged.

  According to the third aspect of the substrate processing method, the presence / absence of liquid splash can be determined using the first statistic used for the presence / absence of ejection. Therefore, the arithmetic processing is simple.

  According to the fourth aspect of the substrate processing method, the pixel value of the pixel in the liquid splash discharge area set in the direction in which liquid splash is likely to occur is used. Therefore, the presence or absence of the discharge nozzle can be appropriately determined.

  According to the fifth aspect of the substrate processing method, since the liquid splash determination area is not set in the area where liquid splash is unlikely to occur, the processing load can be reduced.

  According to the sixth aspect of the substrate processing method, the length of the liquid columnar processing liquid reflected on the upper surface of the substrate becomes long due to the specular reflection, so that the ejection determination region can be easily set.

  According to the seventh aspect of the substrate processing method, in the captured image, the pattern on the upper surface of the substrate is averaged and made uniform, so that the contour of the processing liquid reflected on the upper surface of the substrate can be highlighted.

  According to the eighth aspect of the substrate processing method, in the captured image, the pattern on the upper surface of the substrate is averaged and made uniform, so that the contour of the processing liquid reflected on the upper surface of the substrate can be highlighted.

  According to the ninth aspect of the substrate processing method, the abnormality can be detected with high accuracy.

  According to the tenth aspect of the substrate processing method, the classification accuracy can be improved because the classification can be performed by removing the influence of the region that is less relevant to the ejection state.

  According to the eleventh aspect of the substrate processing method, the nozzle can be moved to the processing position with high accuracy.

It is a figure which shows a schematic example of a structure of a substrate processing apparatus. It is a top view which shows a schematic example of a structure of a processing unit. It is sectional drawing which shows the schematic example of a structure of a processing unit. It is a figure which shows roughly an example of the captured image acquired by the camera. It is a perspective view which shows roughly an example of a structure of a camera and a camera holding part. It is a flow chart which shows an example of substrate processing. It is a flow chart which shows an example of surveillance processing. It is the figure which expanded a part of picked-up image. 7 is a graph showing an example of the brightness value of a pixel in the ejection determination region. It is a figure which shows an example of a captured image roughly. It is a graph which shows an example of the time change of statistics. It is a flow chart which shows an example of surveillance processing. It is a top view which shows a schematic example of a structure of a processing unit. It is a figure which shows an example of a captured image roughly. It is a top view which shows a schematic example of a structure of a processing unit. It is a functional block diagram which shows an example of an internal structure of a control part roughly.

  Hereinafter, embodiments will be described with reference to the accompanying drawings. It should be noted that the drawings are schematically shown, and for convenience of description, the configuration may be omitted or the configuration may be simplified as appropriate. Further, the mutual relationship between the size and the position of the configuration and the like shown in the drawings is not always exactly described and may be changed appropriately.

  Moreover, in the following description, the same components are denoted by the same reference numerals and illustrated, and their names and functions are also the same. Therefore, detailed descriptions thereof may be omitted to avoid duplication.

<Outline of substrate processing equipment>
FIG. 1 is a diagram showing an overall configuration of the substrate processing apparatus 100. The substrate processing apparatus 100 is an apparatus that supplies a processing liquid to the substrate W to process the substrate W. The substrate W is, for example, a semiconductor substrate. The substrate W has a substantially disc shape.

The substrate processing apparatus 100 can remove the unwanted matter attached to the peripheral edge portion of the substrate W by supplying the processing liquid to the edge portion of the substrate W while rotating the substrate W in the horizontal plane. The width of this peripheral edge portion (width along the radial direction) is, for example, about 0.5 to 3 [mm]. Examples of the unnecessary material include films such as SiO ₂ film, SiN film and polysilicon film, and particles. As a treatment liquid for removing such unnecessary substances, hydrofluoric acid (HF), phosphoric acid (H ₃ PO ₄ ), a mixed solution of ammonia (NH ₃ ) and hydrogen peroxide (H ₂ O ₂ ) (SC- 1) and hydrofluoric nitric acid (a mixed solution of hydrofluoric acid and nitric acid (HNO ₃ )). The substrate processing apparatus 100 removes unnecessary substances by supplying the processing liquid to the end portion of the substrate W while rotating the substrate W. Such processing is also called bevel processing.

  The substrate processing apparatus 100 includes an indexer 102, a plurality of processing units 1 and a main transfer robot 103. The indexer 102 has a function of loading the unprocessed substrate W received from the outside of the apparatus into the apparatus and discharging the processed substrate W to the outside of the apparatus. The indexer 102 mounts a plurality of carriers and includes a transfer robot (both not shown). As the carrier, FOUP (front opening unified pod) or SMIF (Standard Mechanical Inter Face) pod that stores the substrate W in a closed space, or OC (open cassette) that exposes the substrate W to the outside air in the stored state is adopted. You can The transfer robot transfers the substrate W between the carrier and the main transfer robot 103.

  Twelve processing units 1 are arranged in the substrate processing apparatus 100. The detailed arrangement is such that four towers in which three processing units 1 are stacked are arranged so as to surround the main transport robot 103. In other words, the four processing units 1 arranged so as to surround the main transfer robot 103 are stacked in three stages, and FIG. 1 shows one of them. The number of processing units 1 mounted on the substrate processing apparatus 100 is not limited to 12, and may be 8 or 4, for example.

  The main transfer robot 103 is installed in the center of the four towers in which the processing units 1 are stacked. The main transfer robot 103 carries in the unprocessed substrate W received from the indexer 102 into each processing unit 1 and carries out the processed substrate W from each processing unit 1 and passes it to the indexer 102.

<Processing unit>
Next, the processing unit 1 will be described. Hereinafter, one of the twelve processing units 1 mounted on the substrate processing apparatus 100 will be described, but the same applies to the other processing units 1. FIG. 2 is a plan view of the processing unit 1. Further, FIG. 3 is a vertical cross-sectional view of the processing unit 1.

  The processing unit 1 includes, as a main element in the chamber 10, a substrate holding unit 20 that holds the substrate W in a horizontal posture (a posture in which a normal line of the substrate W is along the vertical direction), and a substrate held by the substrate holding unit 20. Three processing liquid supply units 30, 60, 65 for supplying the processing liquid to the upper surface of W, a processing cup (cup member) 40 surrounding the periphery of the substrate holding unit 20, and a camera 70 are provided. Further, a partition plate 15 is provided around the processing cup 40 in the chamber 10 to vertically partition the inner space of the chamber 10. Further, the processing unit 1 is provided with a control unit 9 and a notification unit 93.

<Chamber>
The chamber 10 includes a side wall 11 along the vertical direction, a ceiling wall 12 that closes the upper side of a space surrounded by the side wall 11, and a floor wall 13 that closes the lower side. A space surrounded by the side wall 11, the ceiling wall 12, and the floor wall 13 is a processing space for the substrate W. Further, a part of the side wall 11 of the chamber 10 is provided with a loading / unloading port for the main transfer robot 103 to load / unload the substrate W into / from the chamber 10 and a shutter for opening / closing the loading / unloading port (both shown in the drawings). Omitted).

  A fan filter unit (FFU) 14 is attached to the ceiling wall 12 of the chamber 10 for further cleaning the air in the clean room in which the substrate processing apparatus 100 is installed and supplying it to the processing space in the chamber 10. . The fan filter unit 14 includes a fan and a filter (for example, a HEPA filter) for taking in the air in the clean room and sending the air into the chamber 10, and forms a downflow of clean air in the processing space in the chamber 10. In order to uniformly disperse the clean air supplied from the fan filter unit 14, a punching plate having a large number of blowout holes may be provided directly below the ceiling wall 12.

<Substrate holding part>
The substrate holding unit 20 is, for example, a spin chuck. The substrate holding unit 20 includes a disk-shaped spin base 21 fixed in a horizontal posture on an upper end of a rotating shaft 24 extending in the vertical direction. Below the spin base 21, a spin motor 22 that rotates a rotating shaft 24 is provided. The spin motor 22 rotates the spin base 21 in a horizontal plane via a rotation shaft 24. A cylindrical cover member 23 is provided so as to surround the spin motor 22 and the rotary shaft 24.

  The outer diameter of the disk-shaped spin base 21 is slightly larger than the diameter of the circular substrate W held by the substrate holder 20. Therefore, the spin base 21 has a holding surface 21a that faces the entire lower surface of the substrate W to be held.

  A plurality of (four in this embodiment) chuck pins 26 are provided upright on the peripheral portion of the holding surface 21 a of the spin base 21. The plurality of chuck pins 26 are evenly spaced along the circumference corresponding to the outer circumference circle of the circular substrate W (if there are four chuck pins 26 as in this embodiment, at 90 ° intervals). ) It is arranged. The plurality of chuck pins 26 are driven in conjunction with each other by a link mechanism (not shown) housed in the spin base 21. The substrate holding unit 20 holds the substrate W by bringing each of the plurality of chuck pins 26 into contact with the outer peripheral edge of the substrate W to hold the substrate W above the spin base 21 in a horizontal position close to the holding surface 21 a. Can be held (see FIG. 3), and the gripping can be released by separating each of the plurality of chuck pins 26 from the outer peripheral edge of the substrate W.

  When the substrate holding unit 20 holds the substrate W by being gripped by the plurality of chuck pins 26, the spin motor 22 rotates the rotating shaft 24 so that the rotating shaft CX is rotated along the vertical direction passing through the center of the substrate W. The substrate W can be rotated. Here, the substrate holding unit 20 is assumed to rotate counterclockwise in FIG.

<Treatment liquid supply unit>
The processing liquid supply unit 30 includes a discharge nozzle 31, a fixed member 32, and a moving mechanism 33. The fixing member 32 is a member that fixes the discharge nozzle 31, and includes, for example, a nozzle arm 321 and a nozzle base 322. The discharge nozzle 31 is attached to the tip of the nozzle arm 321. The base end side of the nozzle arm 321 is fixedly connected to the nozzle base 322. The moving mechanism 33 moves the discharge nozzle 31 by displacing the fixing member 32. For example, the moving mechanism 33 is a motor, and rotates the nozzle base 322 around an axis along the vertical direction. As the nozzle base 322 rotates, the discharge nozzle 31 is positioned between the processing position above the edge of the substrate W and the standby position outside the processing cup 40, as indicated by an arrow AR34 in FIG. Moves in an arc along the horizontal direction.

  The treatment liquid supply unit 30 may include a plurality of ejection nozzles 31. In the examples of FIGS. 2 and 3, three ejection nozzles 31 are shown as the ejection nozzles 31. The three discharge nozzles 31 are fixed to a nozzle base 322 via a nozzle arm 321. Therefore, the three discharge nozzles 31 move in synchronization with each other. The three discharge nozzles 31 are provided at positions that are arranged along the circumferential direction of the substrate W at the processing position. The distance between the three discharge nozzles 31 in the circumferential direction is, for example, about ten and several [mm].

  As illustrated in FIG. 3, the discharge nozzle 31 is connected to a processing liquid supply source 37 via a pipe 34. An on-off valve 35 is provided in the middle of the pipe 34. A discharge port (not shown) is formed on the lower surface of the tip of the discharge nozzle 31. When the opening / closing valve 35 is opened, the processing liquid from the processing liquid supply source 37 flows through the inside of the pipe 34 and is discharged from the discharge port of the discharge nozzle 31. The processing liquid discharged while the discharge nozzle 31 is stopped at the processing position reaches the edge of the upper surface of the substrate W held by the substrate holding unit 20. As the substrate W rotates, the processing liquid from the discharge nozzle 31 is supplied to the entire area of the peripheral edge portion of the substrate W, and the unwanted matter at the peripheral edge portion is removed (bevel processing).

  A suck back valve 36 may be provided in the middle of each of the pipes 34. The suck back valve 36 draws the processing liquid from the tip of the discharge nozzle 31 by sucking the processing liquid in the pipe 34 when the processing liquid is stopped. As a result, when the discharge is stopped, the processing liquid does not easily drop from the tip of the discharge nozzle 31 as a relatively large lump (droplet).

  When a plurality of ejection nozzles 31 are provided, the ejection nozzles 31 may be connected to different treatment liquid supply sources 37. That is, the processing liquid supply unit 30 may be configured to supply a plurality of types of processing liquids. Alternatively, at least two of the plurality of ejection nozzles 31 may supply the same processing liquid.

  Further, the processing unit 1 of the present embodiment is further provided with two processing liquid supply units 60 and 65 in addition to the processing liquid supply unit 30 described above. The processing liquid supply units 60 and 65 of this embodiment have the same configuration as the processing liquid supply unit 30 described above. That is, the processing liquid supply unit 60 has a discharge nozzle 61, a fixed member 62, and a moving mechanism 63. Like the fixing member 32, the fixing member 62 has a nozzle arm 621 and a nozzle base 622. The discharge nozzle 61 is attached to the tip of the nozzle arm 621, and the nozzle base 622 is connected to the base end thereof. The moving mechanism 63 is, for example, a motor, and by rotating the nozzle base 622, the ejection nozzle 61 is positioned outside the processing position above the edge of the substrate W and the processing cup 40 as indicated by an arrow AR64. Move it in an arc between the standby position and. The discharge nozzle 61 also supplies the processing liquid to the end portion of the substrate W. As the substrate W rotates, the processing liquid from the discharge nozzle 61 is supplied to the entire area of the peripheral edge portion of the substrate W, and the unwanted matter at the peripheral edge portion is removed (bevel processing).

  The processing liquid supply unit 65 has a discharge nozzle 66, a fixing member 67, and a moving mechanism 68. The fixing member 67 has a nozzle arm 671 and a nozzle base 672. The discharge nozzle 66 is attached to the tip of the nozzle arm 671, and the nozzle base 672 is connected to the base end thereof. The moving mechanism 68 is, for example, a motor, and by rotating the nozzle base 672, the discharge nozzle 66 is positioned outside the processing position above the processing center above the substantially center of the substrate W as indicated by an arrow AR69. Move it in an arc between the standby position and. The discharge nozzle 61 supplies the processing liquid to the substantially center of the substrate W. As the substrate W rotates, the processing liquid from the ejection nozzle 66 spreads from the center of the substrate W and is scattered from the periphery of the substrate W to the outside. This allows the treatment liquid to act on the entire upper surface of the substrate W.

  Each of the processing liquid supply units 60 and 65 may be configured to supply a plurality of types of processing liquids. Alternatively, each of the processing liquid supply units 60 and 65 may be configured to supply a single processing liquid.

  The processing liquid supply units 60 and 65 discharge the processing liquid onto the upper surface of the substrate W held by the substrate holding unit 20 with the discharge nozzles 61 and 66 located at the processing positions. At least one of the processing liquid supply units 60 and 65 mixes a cleaning liquid such as pure water with a pressurized gas to generate droplets, and jets a mixed fluid of the droplets and gas onto the substrate W. It may be a two-fluid nozzle. Further, the number of treatment liquid supply units provided in the treatment unit 1 is not limited to three, and may be one or more. Each of the discharge nozzles of the processing liquid supply units 60 and 65 is also connected to the processing liquid supply source via a pipe like the processing liquid supply unit 30, and an opening / closing valve is provided in the middle of the pipe, and a suck back valve is also provided. May be provided. The bevel processing using the processing liquid supply unit 30 will be described below as a representative.

<Treatment cup>
The processing cup 40 is provided so as to surround the substrate holding unit 20. The processing cup 40 includes an inner cup 41, a middle cup 42, and an outer cup 43. The inner cup 41, the middle cup 42, and the outer cup 43 are provided so as to be able to move up and down. Specifically, the processing unit 1 is provided with an elevating mechanism 44, and the elevating mechanism 44 can raise and lower the inner cup 41, the middle cup 42, and the outer cup 43 individually. The elevating mechanism 44 has, for example, a ball screw mechanism.

  When the inner cup 41, the middle cup 42, and the outer cup 43 are elevated, the upper end of the processing cup 40 (here, the upper end of the outer cup 43) is located above the upper surface of the substrate W. Hereinafter, the height position of the upper end of the outer cup 43 when the outer cup 43 is raised is also referred to as the upper end position of the processing cup 40. The vertical interval between the upper end position of the processing cup 40 and the substrate W can be set to about 2 [mm] to 10 [mm], for example.

  When the inner cup 41, the middle cup 42, and the outer cup 43 are elevated, the processing liquid scattered from the peripheral edge of the substrate W hits the inner peripheral surface of the inner cup 41 and falls. The dropped processing liquid is appropriately collected by the first collecting mechanism (not shown). In a state where the inner cup 41 descends and the inner cup 42 and the outer cup 43 rise, the processing liquid scattered from the peripheral edge of the substrate W hits the inner peripheral surface of the middle cup 42 and falls. The dropped processing liquid is appropriately recovered by the second recovery mechanism (not shown). When the inner cup 41 and the middle cup 42 descend and the outer cup 43 rises, the processing liquid scattered from the peripheral edge of the substrate W hits the inner peripheral surface of the outer cup 43 and falls. The processing liquid that has dropped is appropriately collected by a third collecting mechanism (not shown). According to this, different processing liquids can be appropriately collected.

  Hereinafter, the state where the outer cup 43 is raised will be described as the state where the processing cup 40 is raised. That is, when the processing cup 40 is raised, all of the inner cup 41, the middle cup 42, and the outer cup 43 are raised, only the middle cup 42 and the outer cup 43 are raised, and only the outer cup 43 is raised. Including the state of

<Partition board>
The partition plate 15 is provided so as to vertically partition the inner space of the chamber 10 around the processing cup 40. The partition plate 15 may be a single plate-shaped member that surrounds the processing cup 40, or may be a combination of a plurality of plate-shaped members. Further, the partition plate 15 may be formed with a through hole or a notch that penetrates in the thickness direction. In the present embodiment, the nozzle bases 322, 622, 672 of the processing liquid supply units 30, 60, 65 are provided. A through hole (not shown) for passing a support shaft for supporting is formed.

  The outer peripheral end of the partition plate 15 is connected to the side wall 11 of the chamber 10. The edge of the partition plate 15 surrounding the processing cup 40 is formed in a circular shape having a diameter larger than the outer diameter of the outer cup 43. Therefore, the partition plate 15 does not hinder the lifting and lowering of the outer cup 43.

  Further, an exhaust duct 18 is provided in a part of the side wall 11 of the chamber 10 and near the floor wall 13. The exhaust duct 18 is communicatively connected to an exhaust mechanism (not shown). Of the clean air supplied from the fan filter unit 14 and flowing down in the chamber 10, the air that has passed between the processing cup 40 and the partition plate 15 is discharged from the exhaust duct 18 to the outside of the apparatus.

<Camera>
The camera 70 is installed in the chamber 10 and above the partition plate 15. The camera 70 includes, for example, an image pickup element (for example, CCD (Charge Coupled Device)) and an optical system such as an electronic shutter and a lens. The camera 70 can capture an image capturing area described below. That is, the imaging area is an area viewed from an imaging position above the substrate W, and is a substantially liquid column shape ejected from the tip of the ejection nozzle 31 at the processing position and the edge of the substrate W from the tip. This is a region including the treatment liquid of (see also FIG. 3).

  FIG. 4 is a diagram schematically showing an example of image data (hereinafter, referred to as a captured image) IM1 acquired by the camera 70. In the example of FIG. 4, the captured image IM1 includes the tips of the three ejection nozzles 31. The captured image IM1 includes the treatment liquid Lq1 having a substantially liquid column shape ejected from the ejection nozzle 31 located at the center among the three ejection nozzles 31. The substantially liquid columnar processing liquid Lq1 referred to here is the processing liquid Lq1 flowing down from the tip of the discharge nozzle 31 toward the upper surface of the substrate W. The camera 70 outputs this captured image IM1 to the control unit 9.

  As illustrated in FIG. 2, the camera 70 may be movably provided. In the example of FIG. 2, the camera 70 is fixed to the fixing member 62 of the treatment liquid supply unit 60. As a more specific example, a camera holding portion 73 that holds the camera 70 is provided, and the camera holding portion 73 is connected to the nozzle arm 621 of the fixing member 62. For example, the camera holding part 73 is fixed to the tip part of the nozzle arm 621 by a fastening member (for example, a screw) on the base end side, and the camera 70 is fixed and held by the fastening member on the tip side. The camera holding portion 73 is made of, for example, metal (for example, stainless steel) or the like. The moving mechanism 63 moves the camera 70 to the imaging position above the substrate W by displacing the fixed member 62. Specifically, the moving mechanism 63 can reciprocate the camera 70 between the imaging position above the substrate W and the standby position outside the processing cup 40 by rotating the nozzle base 622. it can.

  In the example of FIG. 2, the standby position of the discharge nozzle 31 is at a position deviated from the standby position of the camera 70 by approximately 90 degrees in the clockwise direction. The discharge nozzle 31 and the camera 70 move from their respective standby positions so as to approach each other, and stop at the processing position and the imaging position, respectively. At the image pickup position, the camera 70 is held by the camera holding unit 73 in a posture capable of picking up an image pickup area including the tip of the discharge nozzle 31 and the substantially liquid columnar processing liquid Lq1 discharged from the tip. In the example of FIG. 2, the camera holding portion 73 obliquely projects in the clockwise direction with respect to the nozzle arm 621, and holds the camera 70 at the tip side thereof.

  Here, an example of the positional relationship between the camera 70 and the ejection nozzle 31 when the ejection nozzle 31 is stopped at the processing position and the camera 70 is stopped at the imaging position will be described. Hereinafter, the positional relationship will be described using the ejection nozzle 31 located at the center among the three ejection nozzles 31.

  In the example of FIG. 2, the camera 70 is located on the center side of the substrate W with respect to the ejection nozzle 31 in a plan view. That is, the radial position of the camera 70 with respect to the substrate W is located closer to the center of the substrate W than the radial position of the ejection nozzle 31.

  Further, in the example of FIG. 2, the camera 70 images the tips of the three discharge nozzles 31 from a direction closer to the circumferential direction than the radial direction of the substrate W in a plan view. That is, the position of the camera 70 in the circumferential direction with respect to the substrate W is displaced to one side with respect to the position of the ejection nozzle 31 in the circumferential direction. In other words, in a plan view, an angle θ1 (0 <θ1 <90) formed by the virtual straight line L1 connecting the center of the substrate W and the ejection nozzle 31 and the optical axis of the camera 70 is a virtual line orthogonal to the straight line L1. Is larger than an angle θ2 (0 <θ2 <90) formed by the straight line L2 and the optical axis of the camera 70. According to this, in the captured image IM1, it is possible to make it easy to see the radial position of the liquid deposition position of the treatment liquid Lq1 on the substrate W. However, if the angle θ2 is too small, the three ejection nozzles 31 may be arranged side by side in the depth direction as seen from the imaging position. In this case, since it is difficult to include all the three ejection nozzles 31 in the captured image IM1, it is preferable that the angle θ2 be set so that the three ejection nozzles 31 are appropriately displaced laterally when viewed from the imaging position.

  Further, when the camera 70 captures an image of the imaging region from a direction closer to the circumferential direction, the three ejection nozzles 31 are displaced from each other in the depth direction when viewed from the imaging position. The distance between the three ejection nozzles 31 in the depth direction is, for example, about several [mm] to several tens [mm]. The depth of field of the camera 70 is set so large that the contours of these three ejection nozzles 31 are clear. The distance between the camera 70 and the discharge nozzle 31 is, for example, about 100 [mm].

  In the example of FIG. 2, the camera 70 is located upstream of the ejection nozzle 31 in the rotation direction of the substrate holding unit 20. On the upstream side of the discharge nozzle 31, the amount of the processing liquid Lq1 on the peripheral edge of the substrate W may be smaller than that on the downstream side. This is because the processing liquid Lq1 can be scattered outward from the peripheral edge of the substrate W as the substrate W rotates. Therefore, if the camera 70 is located upstream of the ejection nozzle 31, the treatment liquid Lq1 may adhere to the camera 70, or the vaporized component of the treatment liquid Lq1 may affect the camera 70. Hateful. That is, it is preferable that the camera 70 is located upstream of the discharge nozzle 31 from the viewpoint of protecting the camera 70.

  Now, when the discharge nozzle 31 discharges the processing liquid Lq1, the processing cup 40 is in a raised state. This is because the processing liquid Lq1 scattered from the peripheral edge of the substrate W is received by the processing cup 40. In this state, the tip (ejection port) of the ejection nozzle 31 is located lower than the upper end position of the processing cup 40. For example, the vertical distance between the upper end position of the processing cup 40 and the upper surface of the substrate W is set to approximately 2 [mm] to 10 [mm], and the distance between the discharge nozzle 31 and the substrate W is It is set to about 2 [mm] or less (for example, about 1 [mm]).

  Here, for comparison, a case where the imaging position of the camera 70 is set outside the processing cup 40 will be described. For example, the imaging position is set on the side closer to the discharge nozzle 31 in the space outside the processing cup 40 (the upper right area in the chamber 10 in FIG. 3). Since the upper end position of the processing cup 40 is higher than the tip of the ejection nozzle 31, the processing cup 40 may hinder the imaging. That is, even if an attempt is made to capture an image of the substantially liquid columnar processing liquid Lq1 from an imaging position outside the processing cup 40, the processing liquid Lq1 can be blocked by the processing cup 40. If the imaging position is set to a higher position to avoid the processing cup 40, the ejection nozzle 31 is imaged obliquely from above. Since the distance between the tip of the discharge nozzle 31 and the substrate W is narrow, when the substantially liquid columnar processing liquid Lq1 is to be imaged obliquely from above, this processing liquid Lq1 can be blocked by the discharge nozzle 31 this time.

  Therefore, it is conceivable to set the imaging position in the space outside the processing cup 40 on the side opposite to the discharge nozzle 31 with respect to the center of the substrate W (upper left area in the chamber 10 in FIG. 3). Thereby, it may be possible to capture an image of the substantially liquid-columnar processing liquid Lq1 discharged from the discharge nozzle 31. However, since the distance between the tip of the discharge nozzle 31 and the imaging position of the camera 70 becomes long, a high resolution camera 70 or a telephoto camera 70 is required.

  On the other hand, in the present embodiment, since the image pickup position is above the substrate W, it is easy to bring the image pickup position closer to the upper surface of the substrate W in the height direction, and the optical axis of the camera 70 is horizontal. Easy to follow the direction. Therefore, the camera 70 can image the substantially liquid columnar processing liquid Lq1 discharged from the discharge nozzle 31 without being blocked by the processing cup 40 and the discharge nozzle 31. The angle formed by the optical axis of the camera 70 and the horizontal plane may be set to, for example, about ten degrees [degrees] or less.

  It is also possible to bring the camera 70 close to the ejection nozzle 31 in a plan view. Therefore, it is possible to employ a cheaper camera that has a lower resolution or does not require telephoto. The size of such a camera is small, which is preferable. In the example of FIG. 4, since the distance between the camera 70 and the ejection nozzle 31 is short, the captured image IM1 includes only a part of the peripheral edge of the substrate W.

  Here, an example of the imaging position in the height direction of the camera 70 will be described. The imaging position of the camera 70 may be set so that the lower end of the light receiving surface of the imaging element of the camera 70 is the same as the upper end position of the processing cup 40 or lower than the upper end position. For example, the distance between the camera 70 and the upper surface of the substrate W can be set to about 1 [mm] to 5 [mm]. As a result, the camera 70 can be brought closer to the upper surface of the substrate W, and the optical axis of the camera 70 can be aligned more horizontally.

  Alternatively, the imaging position of the camera 70 may be set such that the lower end of the housing of the camera 70 is the same as or lower than the upper end position of the processing cup 40.

  Further, the camera holding portion 73 may support the lower surface of the camera 70. FIG. 5 is a perspective view schematically showing an example of the camera 70 and the camera holding unit 73. In FIG. 5, the substrate W and the ejection nozzle 31 are also shown. In the example of FIG. 5, the camera holding portion 73 includes an L-shaped connecting member 731, an upper surface member 732 located on the upper surface side of the camera 70, a side surface member 733 located on the side surface side of the camera 70, and a camera 70. And a lower surface member 734 located on the lower surface side. The connecting member 731 has a first rod-shaped member that extends horizontally from the nozzle arm 621 and a second rod-shaped member that extends vertically downward from the tip of the rod-shaped member. The tip of the second rod-shaped member is connected to the upper surface member 732. In the example of FIG. 5, the upper surface member 732, the side surface member 733, and the lower surface member 734 have a plate shape. The upper surface member 732 and the lower surface member 734 are arranged such that their thickness direction is along the vertical direction, and the side surface member 733 is arranged such that their thickness direction is along the horizontal direction. The side surface member 733 connects the upper surface member 732 and the lower surface member 734. The lower surface member 734 also functions as a support member that supports the camera 70.

  In such a structure, the imaging position of the camera 70 may be set such that the lower end of the lower surface member 734 is the same as the upper end position of the processing cup 40 or lower than the upper end position. Also by this, the camera 70 can be brought closer to the upper surface of the substrate W, and the optical axis of the camera 70 can be further aligned in the horizontal direction.

<Lighting unit>
As shown in FIG. 3, an illumination section 71 is provided in the chamber 10 and above the partition plate 15. The illumination unit 71 includes a light source such as an LED (Light Emitting Diode). The wavelength of light emitted by the illumination unit 71 is not particularly limited, but visible light or near infrared light can be used, for example. In the example of FIG. 3, the illumination unit 71 is arranged above the camera 70. For example, the illumination unit 71 is arranged at a position overlapping the camera 70 in plan view (see FIG. 2). The illumination unit 71 may be held by the camera holding unit 73. For example, the illumination section 71 may be fixed to the upper surface of the upper surface member 732 of the camera holding section 73. Since the inside of the chamber 10 is usually a dark room, the illumination unit 71 irradiates the imaging area with light when the camera 70 performs imaging.

<Control part>
The control unit 9 controls various components of the substrate processing apparatus 100 and advances the processing on the substrate W. The control unit 9 also performs image processing on the captured image IM1 acquired by the camera 70. Therefore, the control unit 9 functions as an image processing unit. Since the camera 70 images the tip of the ejection nozzle 31 from the imaging position above the substrate W, the substantially liquid columnar processing liquid Lq1 ejected from the ejection nozzle 31 is appropriately included in the captured image IM1 acquired by the camera 70. include. The control unit 9 monitors the ejection state of the treatment liquid Lq1 ejected from the ejection nozzle 31 by performing image processing on the captured image IM1 (bevel monitoring). An example of this monitoring process will be described later in detail.

  The hardware configuration of the control unit 9 is similar to that of a general computer. That is, the control unit 9 stores a CPU that performs various arithmetic processes, a ROM that is a read-only memory that stores a basic program, a RAM that is a readable / writable memory that stores various information, and control software and data. It is equipped with a magnetic disk and the like. When the CPU of the control unit 9 executes a predetermined processing program, each operating mechanism of the substrate processing apparatus 100 is controlled by the control unit 9, and the processing in the substrate processing apparatus 100 proceeds. The CPU of the control unit 9 executes a predetermined processing program to perform image processing. Note that some or all of the functions of the control unit 9 may be realized by dedicated hardware.

<Notification section>
The notification unit 93 is, for example, a voice output unit (for example, a speaker) or a display. The notification unit 93 can give various notifications to the operator. For example, the voice output unit outputs a notification sound (buzzer or voice), or the display displays the notification information, whereby various notifications can be given to the worker. The notification of the notification unit 93 is controlled by the control unit 9.

<Operation of control unit>
FIG. 6 is a flowchart showing an example of the substrate processing. First, in step S1, the main transport robot 103 transports the substrate W onto the substrate holder 20. The substrate holding unit 20 holds the transported substrate W.

  Next, in step S2, the control unit 9 controls the moving mechanism 33 to move the discharge nozzle 31 to the processing position, and controls the moving mechanism 63 to move the camera 70 to the imaging position. Next, in step S3, the control unit 9 controls the elevating mechanism 44 to raise the processing cup 40, and controls the spin motor 22 to rotate the spin base 21. The rotation speed of the spin base 21 is set to, for example, about 1000 [rpm] or higher.

  Next, in step S4, the control unit 9 controls the camera 70 to start imaging. The camera 70 images the imaging area at a predetermined frame rate (for example, 60 frames / second), and sequentially outputs the acquired captured image IM1 to the control unit 9. The control unit 9 monitors the ejection state of the processing liquid Lq1 based on the image processing on the captured image IM1 as described later in detail.

  Next, in step S5, the control unit 9 starts the ejection of the treatment liquid Lq1 from the ejection nozzle 31. Specifically, the control unit 9 outputs an open signal to the open / close valve 35. The on-off valve 35 performs an opening operation based on this open signal to open the pipe 34. As a result, the processing liquid Lq1 from the processing liquid supply source 37 is ejected from the ejection nozzle 31 and reaches the end portion of the upper surface of the substrate W. The flow rate of the treatment liquid Lq1 is set to, for example, about several to several tens [ml / min]. This flow rate is smaller than the flow rate of the processing liquid when processing the entire surface of the substrate W (for example, the flow rate of the processing liquid discharged from the discharge nozzle 66 of the processing liquid supply unit 65).

  The processing liquid Lq1 is discharged to the end portion of the substrate W while rotating the substrate W, so that the processing liquid Lq1 acts on the entire region of the peripheral edge portion of the substrate W. The processing liquid Lq1 can remove unnecessary substances attached to the peripheral edge portion of the substrate W (bevel processing). From the discharge ports of the three discharge nozzles 31, the processing liquid Lq1 according to the type of the unnecessary substance (for example, a film) can be sequentially discharged. The treatment liquid may be ejected from at least two ejection ports of the three ejection nozzles 31 at the same time.

  Since the flow rate of the processing liquid Lq1 is small in this bevel processing, the processing liquid Lq1 is easily affected by the air flow accompanying the rotation of the substrate W, and as a result, the liquid splashing the processing liquid Lq1 on the upper surface of the substrate W is likely to occur.

  Therefore, the control unit 9 monitors the discharge state of the processing liquid Lq1 in the monitoring process. The specific operation of the monitoring process will be described later in detail.

  When the bevel processing end condition is satisfied, the control unit 9 stops the discharge of the processing liquid Lq1 from the discharge nozzle 31 in step S6. Although the bevel processing termination condition is not particularly limited, for example, a condition that the elapsed time from step S5 reaches a predetermined time can be adopted. The control unit 9 outputs a close signal to the on-off valve 35 in response to the satisfaction of this ending condition. The on-off valve 35 performs a closing operation based on this open signal to close the pipe 34. As a result, the discharge of the processing liquid Lq1 is completed. When the suck back valve 36 is provided, the controller 9 outputs a suction signal to the suck back valve 36.

  The step of drying the substrate W may be appropriately performed after the ejection of the processing liquid Lq1 is stopped. Next, in step S7, the control unit 9 causes the camera 70 to finish imaging. That is, the monitoring process ends. Next, in step S8, the control unit 9 controls the spin motor 22 to end the rotation of the spin base 21, and controls the elevating mechanism 44 to lower the processing cup 40. Next, in step S9, the control unit 9 controls the moving mechanism 33 and the moving mechanism 63 to move the ejection nozzle 31 and the camera 70 to their respective standby positions.

  FIG. 7 is a flowchart showing an example of the operation of the monitoring process. The processing flow shown in FIG. 7 is executed, for example, every time the captured image IM1 is input to the control unit 9. First, in step S11, the control unit 9 specifies an ejection determination region R2 described below in the captured image IM1.

  FIG. 8 is a diagram schematically showing an example of an enlarged view of the captured image IM1. In the example of FIG. 8, an enlarged view of the region R1 near the tip of one ejection nozzle 31 is shown. The ejection determination region R2 is a region immediately below the ejection nozzle 31 in the captured image IM1, and is a region including a part of the substantially liquid columnar processing liquid Lq1 ejected from the ejection nozzle 31. The ejection determination region R2 is set at a position apart from the ejection nozzle 31 in the captured image IM1. The ejection determination region R2 has a long shape that is long in the lateral direction. That is, the width in the vertical direction of the ejection determination region R2 is narrower than the width in the horizontal direction. More specifically, the lateral width of the ejection determination region R2 is wider than the liquid injection width of the processing liquid Lq1 ejected from the ejection nozzle 31, and is set to, for example, three times or more the normal liquid column width. The lateral position of the ejection determination region R2 is set so that both ends in the width direction of the treatment liquid Lq1 are included in the ejection determination region R2. The vertical width of the ejection determination region R2 is appropriately set, and for example, the width may be a width corresponding to several pixels.

  The ejection determination region R2 in the captured image IM1 is preset for the ejection nozzle 31. That is, the relative positional relationship between the ejection nozzle 31 and the ejection determination region R2 is preset. Information indicating this positional relationship may be stored in the storage medium of the control unit 9.

  By the way, since the relative position of the camera 70 with respect to the discharge nozzle 31 can change according to the accuracy of the moving mechanisms 33 and 63, the position of the discharge nozzle 31 in the captured image IM1 can also change. Therefore, the control unit 9 may identify the position of the ejection nozzle 31 in the captured image IM1 and identify the ejection determination region R2 that has a predetermined positional relationship with the identified ejection nozzle 31. A reference image including the appearance of the tip of the ejection nozzle 31 is also stored in advance in the storage medium of the control unit 9 in order to identify the position of the ejection nozzle 31 in the captured image IM1. The control unit 9 identifies the position of the ejection nozzle 31 in the captured image IM1 by pattern matching based on the reference image, and the ejection determination region R2 based on a predetermined relative positional relationship with the identified ejection nozzle 31. Specify. Accordingly, even if the position of the ejection nozzle 31 changes in the captured image IM1, the ejection determination region R2 can be appropriately specified in accordance with the position of the ejection nozzle 31.

  When the ejection nozzle 31 is ejecting the treatment liquid Lq1, the ejection determination region R2 includes a part of the treatment liquid Lq1 having a substantially liquid column shape. Since the light emitted by the illumination unit 71 is reflected by the treatment liquid Lq1 and received by the camera 70, the luminance value of the pixel reflecting the treatment liquid Lq1 becomes higher than the luminance values of the other pixels. When the camera 70 is a grayscale monochrome camera, it can be said that the pixel value of the pixel indicates the brightness value. Here, as an example, the camera 70 is assumed to be a monochrome camera.

  FIG. 9 is a graph showing an example of luminance values (pixel values in this case) of pixels in the ejection determination region R2. The horizontal axis represents the pixel number of pixels lined up in a row in the ejection determination region R2, and the vertical axis represents the pixel value of pixels lined up in a row in the ejection determination region R2. As illustrated in FIG. 9, the brightness corresponding to the liquid column portion of the processing liquid Lq1 is higher than the surroundings. That is, the brightness distribution has a feature due to the liquid column shape of the treatment liquid Lq1.

  Returning to FIG. 7, next, in step S12, the control unit 9 calculates the statistical amount A1 of the pixel value of the pixel in the identified ejection determination region R2. This statistic A1 is an amount that reflects the discharge state of the treatment liquid Lq1. As the statistic A1, for example, a variance (for example, standard deviation) of pixel values in the ejection determination region R2 can be adopted. This is because the pixel value of some pixels (pixels corresponding to the treatment liquid Lq1) in the ejection determination region R2 increases due to the ejection of the treatment liquid Lq1 (see FIG. 9), and thus the treatment liquid Lq1 is ejected. This is because it increases compared to when it is not present. That is, it can be said that the dispersion is a value that reflects whether or not the processing liquid Lq1 is ejected.

  The dispersion is also a value that reflects whether or not the processing liquid Lq1 ejected from the ejection nozzle 31 splashes on the upper surface of the substrate W. The reason will be described below. FIG. 10 is a diagram schematically showing an example of the captured image IM1 when liquid splash occurs. As illustrated in FIG. 10, when the liquid splashes, the region occupied by the treatment liquid Lq1 in the ejection determination region R2 expands. That is, in the ejection determination region R2, a liquid splash portion is included beside the liquid column portion of the treatment liquid Lq1, and the region occupied by the treatment liquid Lq1 is expanded as a whole. In the liquid splash portion of the treatment liquid Lq1, the variation in the distribution of the luminance value is larger than in the liquid column portion, and as a result, the dispersion of the pixel values in the ejection determination region R2 is further increased.

  FIG. 11 is a graph showing an example of the change over time of the statistic A1. The horizontal axis represents time. As the parameter indicating the time, for example, the frame number in the captured image IM1 may be adopted. The vertical axis represents the statistic A1. Here, the standard deviation is adopted as the statistic A1. In FIG. 11, three graphs G1 to G3 having different discharge times of the treatment liquid Lq1 are shown. In the example of FIG. 11, the ejection time of the graph G1 is the longest, and the ejection time of the graph G3 is the shortest. Graphs G1 to G3 are graphs when the flow rates of the treatment liquid Lq1 are 18 [ml / min], 12 [ml / min], and 8 [ml / min], respectively.

  The graph G1 and the graph G2 show the time change of the statistical amount A1 when the treatment liquid Lq1 is appropriately discharged. The graph G3 shows the time change of the statistic A1 when the liquid splash occurs immediately after the start of the discharge of the treatment liquid Lq1. When the ejection nozzle 31 is not ejecting the treatment liquid Lq1, the statistic amount A1 is less than the threshold value Aref1. When the ejection nozzle 31 ejects the treatment liquid Lq1, the statistic amount A1 increases in accordance with the ejection and exceeds the threshold value Aref1. When the ejection nozzle 31 is ejecting the processing liquid Lq1 normally, the statistic amount A1 is less than the threshold value Aref2. The threshold Aref2 is larger than the threshold Aref1. The threshold Aref1 and the threshold Aref2 can be set in advance by simulation, experiment, or the like, and may be stored in the storage medium of the control unit 9, for example.

  On the other hand, when the liquid splashes, the statistic A1 exceeds the threshold Aref2. In the example of FIG. 11, liquid splashing occurs immediately after the start of discharge of the processing liquid Lq1 having the smallest flow rate (see graph G3). This can be considered as follows. That is, it can be considered that the processing liquid Lq1 having a small flow rate is easily affected by the air flow accompanying the rotation of the substrate W, and the flow of the processing liquid Lq1 is not stable immediately after the start of ejection.

  In the example of FIG. 11, the statistic A1 greatly increases according to the liquid splash. This can be considered for the following reasons.

  That is, in the bevel process, since the flow rate of the treatment liquid Lq1 is small and the liquid column of the treatment liquid Lq1 is thin, the ratio of the area occupied by the liquid splash portion to the area occupied by the liquid column portion in the discharge determination region R2 is compared. Will grow larger. In other words, in the bevel processing, the liquid splash portion having a large variation in luminance distribution occupies a relatively large amount.

  For comparison, the processing liquid supply unit 65 will be described. In the processing liquid supply unit 65, since the flow rate of the processing liquid discharged from the discharge nozzle 66 is large, the liquid injection of the processing liquid is thick. Therefore, even if the liquid splash occurs, the ratio of the liquid splash portion to the liquid column portion in the discharge determination region does not increase as much as the bevel process.

  Moreover, in the processing liquid supply unit 65, the distance between the discharge nozzle 66 and the substrate W is set to be wide. This is because the flow rate is high. In this case, the discharge determination region is set longer in the vertical direction than in the horizontal direction in order to capture the discharge state of the treatment liquid Lq1 in a wider range. Then, when the camera images the tip of the ejection nozzle 66 from a direction close to horizontal, the vertical width of the liquid splash portion of the treatment liquid Lq1 appears relatively narrow in the captured image IM1. Therefore, the liquid splash portion of the processing liquid Lq1 exists only in the lower portion of the ejection determination region. In this case, the ratio of the liquid splash portion to the liquid column portion in the discharge determination region is not so large.

  On the other hand, in the bevel process, the interval between the tip of the ejection nozzle 31 and the upper end of the substrate W is narrow, and the ejection determination region R2 is set to be horizontally long. Therefore, the liquid splash portion of the treatment liquid Lq1 may exist from the end to the end of the ejection determination region R2 in the vertical direction of the captured image IM1 (see also FIG. 10).

  As described above, by setting the horizontally long ejection determination region R2 in the bevel process, when liquid splash occurs, the proportion of the liquid splash portion in the ejection determination region R2 increases significantly. Therefore, the occurrence of the liquid splash significantly increases the dispersion in the ejection determination region R2. That is, because the horizontally long ejection determination region R2 is set in the bevel processing, the dispersion easily reflects the presence or absence of liquid splash.

  Referring to FIG. 7, next, in step S13, control unit 9 determines whether statistic A1 is equal to or greater than threshold value Aref1. When the statistic amount A1 is less than the threshold value Aref1, the control unit 9 determines in step S14 that the ejection state is the ejection stop state, and ends the process. That is, the control unit 9 determines that the processing liquid Lq1 has not been discharged yet, and ends the processing.

  When the statistic A1 is greater than or equal to the threshold Aref1, the control unit 9 determines in step S15 whether the statistic A1 is less than the threshold Aref2. When the statistic amount A1 is less than the threshold value Aref2, in step S16, the control unit 9 determines that the ejection state is the normal ejection state, and ends the process. That is, the control unit 9 determines that the processing liquid Lq1 is normally discharged, and ends the processing.

  When the statistic amount A1 is equal to or larger than the threshold value Aref2, in step S17, the control unit 9 determines that the ejection state is the liquid splashing state, and ends the process. That is, the control unit 9 determines that liquid splash has occurred, and ends the process. When it is determined that the liquid splash has occurred, the control unit 9 may notify the notification unit 93 that the liquid splash has occurred. According to this, the operator can recognize that the liquid splash has occurred.

  As described above, according to the processing unit 1, it is possible to appropriately determine the ejection state (presence / absence of ejection and presence / absence of liquid splash) of the treatment liquid Lq1 in the bevel processing. Moreover, since the presence or absence of liquid splash can be determined using the statistic A1 used for the presence or absence of ejection, the arithmetic processing is simple.

  In the above example, the variance is used as the statistic A1, but the present invention is not limited to this. As the statistic amount A1, the sum (or average, the same applies below) of the pixel values of the pixels in the ejection determination region R2 can be adopted. This is because the pixel value of the pixel corresponding to the treatment liquid Lq1 is higher than the pixel values of the other pixels in the ejection determination region R2 (see also FIG. 9). That is, the total sum increases due to the ejection of the processing liquid Lq1. In other words, the total sum is a value that reflects the presence / absence of ejection of the treatment liquid Lq1.

  The total sum is also a value that reflects the presence or absence of liquid splash. The reason will be described below. As illustrated in FIG. 10, when the liquid splashes, the region occupied by the treatment liquid Lq1 in the ejection determination region R2 expands. That is, the number of pixels having a high brightness value in the ejection determination region R2 further increases, and the total sum of pixel values in the ejection determination region R2 further increases.

  The tendency of the change of the total depending on the presence / absence of the ejection and the presence / absence of liquid splashing is similar to the graph of FIG. 11. Therefore, the total sum can be adopted as the statistic A1.

  Note that it can be considered that the reason why the total sum greatly increases depending on the presence / absence of the liquid splash is that the horizontally long ejection determination region R2 is set in the bevel processing, similarly to the dispersion. That is, in the horizontally long ejection determination region R2 in the bevel process, the proportion of the liquid splash portion to the liquid column portion is large. Therefore, the total sum of the pixel values in the ejection determination region R2 is significantly increased compared to the case where the processing liquid Lq1 is ejected normally. That is, in the bevel processing, the total sum easily reflects the presence or absence of the splash of the processing liquid Lq1.

<Liquid splash judgment area>
In the above example, the presence or absence of liquid splash is determined based on the pixel value of the pixel in the ejection determination region R2. However, a liquid splash determination region R3 different from the ejection determination region R2 may be set. Similar to the ejection determination area R2, the liquid splash determination area R3 is set in advance corresponding to the position of the ejection nozzle 31 in the captured image IM1. That is, the relative position between the ejection nozzle 31 and the liquid splash determination region R3 in the captured image IM1 is set in advance. Information indicating this relative position is stored in, for example, the storage medium of the control unit 9.

  In the example of FIGS. 8 and 11, the liquid splash determination region R3 is set apart from the ejection determination region R2 in the captured image IM1. More specifically, the liquid splash determination region R3 is set downstream of the ejection determination region R2 in the rotational direction. In other words, the liquid splash determination region R3 does not have to be set on the upstream side in the rotation direction with respect to the discharge determination region R2. This is for the following reason. That is, in the bevel processing, the processing liquid Lq1 is easily affected by the air flow associated with the rotation of the substrate W, and therefore the liquid splash mainly occurs on the downstream side in the rotation direction with respect to the liquid deposition position of the processing liquid Lq1. Therefore, the liquid splash determination region R3 is set in a region where liquid splash is likely to occur, and the liquid splash determination region R3 is not set in a region where liquid splash is less likely to occur.

  In the example of FIGS. 8 and 11, the liquid splash determination region R3 has a rectangular shape, the width in the horizontal direction thereof is narrower than that of the discharge determination region R2, and the width in the vertical direction thereof is the discharge determination region R2. Wider than. According to this, the liquid splash can be more effectively included in the liquid splash determination region R3.

  When the liquid splash does not occur, only the upper surface of the substrate W is included in the liquid splash determination region R3, so that the pixel values of the pixels in the liquid splash determination region R3 are small and the variation in the distribution is small. . On the other hand, when the liquid splash occurs, the liquid splash determination region R3 includes the liquid splash portion of the treatment liquid Lq1, and thus the pixel value of the pixel corresponding to the liquid splash portion increases. To do.

  Therefore, the control unit 9 calculates the statistic B1 based on the pixel values of the pixels in the liquid splash determination region R3. The statistic amount B1 is an amount that reflects the presence or absence of liquid splash, and is, for example, the sum or variance of pixel values in the liquid splash determination region R3. This statistic B1 increases due to the occurrence of liquid splash, similar to the statistic A1.

  FIG. 12 is a flowchart showing an example of the monitoring process. The processing flow shown in FIG. 12 is executed, for example, every time the captured image IM1 is input to the control unit 9. First, in step S21, the control unit 9 specifies the liquid splash determination region R3 in the captured image IM1. Specifically, the control unit 9 specifies the position of the ejection nozzle 31 in the captured image IM1 by pattern matching based on the reference image, and based on a predetermined relative positional relationship with respect to the specified ejection nozzle 31. And specifies the liquid splash determination region R3. Thereby, even if the position of the ejection nozzle 31 changes in the captured image IM1, the liquid splash determination region R3 can be appropriately specified in accordance with the position of the ejection nozzle 31.

  Next, in step S22, the control unit 9 calculates the statistical amount B1 of the pixel value of the pixel in the liquid splash determination region R3. Next, in step S23, the control unit 9 determines whether or not the statistic amount B1 is greater than or equal to the threshold value Bref1. The threshold value Bref1 may be set in advance by, for example, simulation or experiment, and may be stored in the storage medium of the control unit 9.

  When the statistic amount B1 is equal to or greater than the threshold value Bref1, the control unit 9 determines in step S24 that the ejection state is the liquid splashing state, and ends the process. That is, the control unit 9 determines that liquid splash has occurred, and ends the process.

  On the other hand, when the statistic amount B1 is less than the threshold value Bref1, the control unit 9 ends the process without executing step S24.

  As described above, the control unit 9 can appropriately determine the presence or absence of the liquid splash based on the pixel value in the liquid splash determination region R3.

  Further, in the above example, the liquid splash determination region R3 is not set on the upstream side in the rotation direction with respect to the discharge determination region R2. Therefore, since the calculation process for the liquid splash determination region R3 is not performed, the processing load can be reduced.

  In the above example, the variance of the pixel values of all the pixels in the ejection determination region R2 is used as the variance of the statistic A1. However, the pixel values of pixels arranged in the vertical direction may be integrated for each column, and the variance of the plurality of integrated values in the plurality of columns may be adopted. The same applies to the statistic B1.

<Camera fixation>
In the above example, the camera 70 is fixed to the fixing member 62 similarly to the ejection nozzle 61. That is, the mechanism for moving the camera 70 and the mechanism for moving the discharge nozzle 61 are combined. Therefore, the manufacturing cost and size can be reduced as compared with the case where a dedicated mechanism is provided for each.

  FIG. 13 is a plan view schematically showing an example of the configuration of the processing unit 1A. The processing unit 1A has the same configuration as the processing unit 1 except that the camera 70 is fixed. In the processing unit 1A, the camera 70 is fixed to the fixing member 32 in the same manner as the ejection nozzle 31 to be imaged. More specifically, the camera holder 73 is connected to the nozzle arm 321 on the side of the nozzle arm 321. The camera holding unit 73 holds the camera 70. The camera 70 is fixed to the fixing member 32 via the camera holding portion 73. The camera 70 and the camera holding portion 73 are arranged on the counterclockwise direction side (that is, the side from the standby position of the ejection nozzle 31 toward the processing position) with respect to the nozzle arm 321. Further, the camera 70 is held by the camera holding portion 73 in a posture capable of imaging the tip of the discharge nozzle 31 and the treatment liquid Lq1 discharged from the discharge nozzle 31.

  By rotating the nozzle base 322 by the moving mechanism 33, the ejection nozzle 31 and the camera 70 can be moved to the processing position and the imaging position, respectively, while maintaining their positional relationship. The positional relationship between the imaging position of the camera 70 and the processing position of the ejection nozzle 31 is the same as that of the processing unit 1.

  With this processing unit 1A as well, like the processing unit 1, the camera 70 can appropriately image the substantially liquid columnar processing liquid Lq1 ejected from the ejection nozzle 31.

  Further, since the camera 70 is fixed to the same fixing member 32 as the discharge nozzle 31, the camera 70 can be positioned with respect to the discharge nozzle 31 with high accuracy. That is, in the processing unit 1, since the discharge nozzle 31 and the camera 70 are fixed to the nozzle arms 321 and 621 different from each other, in view of the accuracy of the moving mechanisms 33 and 63, there is a gap between the camera 70 and the nozzle arm 321. In the processing unit 1A, the ejection nozzle 31 and the camera 70 are fixed to the same nozzle arm 321, whereas a relatively wide margin needs to be provided. Therefore, the margin between the camera 70 and the nozzle arm 321 is narrower. Can be set. That is, the camera 70 can be brought closer to the nozzle arm 321. According to this, the camera 70 can image the discharge nozzle 31 from a direction closer to the circumferential direction. Therefore, it is easy to identify the radial ejection position of the treatment liquid Lq1 in the captured image IM1.

<Camera protection>
When the treatment liquid Lq1 contains hydrofluoric acid, the lower surface of the housing of the camera 70 or the lower end surface of the lower surface member 734 of the camera holding portion 73 is preferably made of a chemical resistant material. In short, the protective member 74 that protects the camera 70 may be provided on the lower surface side of the camera 70. As the protective member 74, a chemically resistant resin such as a fluororesin such as polytetrafluoroethylene or a vinyl chloride resin or a metal such as stainless steel, which has a high chemical resistance to hydrofluoric acid, can be adopted.

  According to this, it is possible to reduce the possibility that the camera 70 located above the substrate W is corroded by the vaporized component of the treatment liquid Lq1. Therefore, the reliability of the camera 70 can be improved.

<Discharge determination area R2>
FIG. 14 is a diagram schematically showing another example of the region R1 of the captured image IM1. In the captured image IM1 illustrated in FIG. 11, the ejection nozzle 31 is included in the upper surface of the substrate W. This is because the light from the illumination unit 71 is reflected by the ejection nozzle 31 and then mirror-reflected on the upper surface of the substrate W to be received by the light receiving surface of the camera 70. That is, the upper surface of the substrate W functions as a mirror, and the appearance of the ejection nozzle 31 is reflected on the upper surface.

  In such a captured image IM1, the ejection determination region R2 may be set so as to include a part of the substantially liquid columnar processing liquid Lq1 (liquid column portion) that is imaged on the upper surface of the substrate W. That is, in the captured image IM1, the ejection determination region R2 may be set so as to cross the substantially liquid-columnar processing liquid Lq1 that appears on the upper surface of the substrate W.

  The control unit 9 can specify the liquid column width and the ejection position of the treatment liquid Lq1 in the ejection determination region R2 by the image processing similar to the above-described image processing.

  Moreover, in the present embodiment, since the optical axis of the camera 70 is along the horizontal direction, the vertical length of the processing liquid Lq1 reflected on the upper surface of the substrate W in the captured image IM1 is the discharge nozzle 31 and the substrate W. It becomes longer than the length of the treatment liquid Lq1 between and. According to this, it is easy to set the ejection determination region R2 so as to cross the treatment liquid Lq1. Further, the width of the ejection determination region R2 in the vertical direction can be set wider.

  By the way, various patterns such as a metal pattern, a semiconductor pattern, an insulating layer pattern, and a resist pattern may be formed on the upper surface of the substrate W. Therefore, the treatment liquid Lq1 that appears on the upper surface of the substrate W is affected by these patterns. As a result, the contour of the treatment liquid Lq1 may be blurred.

  Therefore, the exposure time of the camera 70 may be set to be longer than the rotation time required for one rotation of the substrate W. According to this, the pattern of the substrate W in the captured image IM1 is averaged and made uniform, so that the contour of the treatment liquid Lq1 in the captured image IM1 can be highlighted. According to this, it is possible to improve the accuracy of identifying the positions of both ends of the treatment liquid Lq1, and consequently it is possible to obtain the liquid column width and the ejection position with higher accuracy.

  Alternatively, the exposure time may be shorter than the rotation time. The controller 9 may integrate or average a plurality of captured images IM1 captured within a predetermined time longer than the rotation time to generate a processed image every predetermined time. In the processed image for each predetermined time, the pattern on the upper surface of the substrate W is averaged and made uniform, so that the contour of the processing liquid Lq1 can be made outstanding.

  Although the ejection determination region R2 has been described in the above example, the same applies to the liquid splash determination region R3.

<Position of discharge nozzle 31>
The processing position of the ejection nozzle 31 may be controlled based on the captured image IM1. Hereinafter, a specific description will be given.

  The processing position of the discharge nozzle 31 is a position separated from the peripheral edge of the substrate W by a predetermined width. Therefore, the control unit 9 specifies the position of the peripheral edge of the substrate W in the captured image IM1 (hereinafter referred to as the substrate peripheral edge position). First, the control unit 9 specifies a peripheral area R4 described below in the captured image IM1.

  The peripheral area R4 is an area including a part of the peripheral edge of the substrate W in the captured image IM1. In the example of FIG. 8, the peripheral region R4 has a rectangular shape. The position of the peripheral region R4 is set in advance corresponding to the position of the ejection nozzle 31, like the ejection determination region R2. That is, the relative positional relationship between the ejection nozzle 31 and the peripheral region R4 is preset. Information indicating this positional relationship may be stored in the storage medium of the control unit 9.

  The control unit 9 specifies the position of the ejection nozzle 31 in the captured image IM1 by pattern matching, and specifies the peripheral region R4 based on the specified position of the ejection nozzle 31. Then, the control unit 9 specifies the substrate peripheral edge position of the substrate W in the peripheral region R4. For example, the control unit 9 identifies the peripheral edge of the substrate W based on image processing such as edge detection processing. Accordingly, the substrate peripheral edge position of the substrate W based on the position of the discharge nozzle 31 can be specified.

  The control unit 9 may determine the processing position of the discharge nozzle 31 based on this substrate peripheral position. For example, when the ejection nozzle 31 stops at the correct processing position, the captured image IM1 is acquired in advance, the substrate peripheral position in the peripheral region R4 is determined in advance in the captured image IM1, and the control unit 9 uses the substrate peripheral position as a reference position in advance. Stored in the storage medium.

  The controller 9 compares the specified substrate peripheral edge position with the reference position and adjusts the position of the discharge nozzle 31 so that the difference is reduced. For example, when the substrate peripheral edge position is displaced leftward from the reference position in the captured image IM1, the control unit 9 controls the moving mechanism 33 to move the discharge nozzle 31 to the center side of the substrate W so that the substrate peripheral edge position is changed. When it is displaced to the right from the reference position, the moving mechanism 33 is controlled to move the discharge nozzle 31 to the peripheral side of the substrate W. As a result, the discharge nozzle 31 can be moved to a position that is the center of the substrate W by a predetermined width than the peripheral position of the substrate.

<Imaging optical system>
FIG. 15 is a diagram schematically showing an example of the configuration of the processing unit 1B. The processing unit 1B has the same configuration as the processing unit 1 except for the imaging optical system. A mirror 75 is provided in the processing unit 1B. The mirror 75 is arranged at the imaging position above the substrate W, and the camera 70 is arranged in a region other than above the substrate W. As illustrated in FIG. 15, the camera 70 may be located above the processing cup 40 in a plan view. The mirror 75 reflects the light from the imaging area toward the light receiving surface of the camera 70. Therefore, the camera 70 can take an image of the imaging region viewed from the imaging position above the substrate W.

  As illustrated in FIG. 15, the mirror 75 may be movably provided. In the example of FIG. 15, the mirror 75 is fixed to the fixing member 62 of the treatment liquid supply unit 60. As a more specific example, a mirror holding portion 76 that holds the mirror 75 is provided, and the mirror holding portion 76 is connected to the nozzle arm 621 of the fixing member 62. For example, the mirror holding portion 76 is fixed to the tip end portion of the nozzle arm 621 at the base end side by a fastening member (for example, a screw), and the mirror 75 is fixedly held at the tip end side by the fastening member. The mirror holding portion 76 is formed of, for example, metal (for example, stainless steel) or the like. By moving the nozzle base 622, the moving mechanism 63 can reciprocate the mirror 75 between the imaging position above the substrate W and the standby position outside the processing cup 40. By moving the mirror 75 to the image pickup position by the moving mechanism 63, the light from the image pickup area can be reflected from the mirror 75 to the camera 70.

  The positional relationship between the position of the mirror 75 (imaging position) and the ejection nozzle 31 in plan view is the same as the positional relationship between the position of the camera 70 (imaging position) and the ejection nozzle 31 in the processing unit 1. The imaging position is preferably close to the substrate W. For example, the imaging position may be set so that the lower end of the reflecting surface of the mirror 75 is the same as the upper end position of the processing cup 40 or lower than the upper end position. . Alternatively, when the mirror holding portion 76 has a lower surface member arranged below the mirror 75, the lower end of the lower surface member is the same as the upper end position of the processing cup 40 or lower than the upper end position. The imaging position may be set so that As a result, the camera 70 can image the imaging region viewed from the imaging position along a direction that is more horizontal. That is, it is easy to make the imaging direction from the imaging position more horizontal.

  According to the processing unit 1B, since the camera 70 can be arranged in the area other than the upper part of the substrate W, the influence of the processing liquid Lq1 on the camera 70 can be reduced. For example, it is possible to reduce the possibility that the treatment liquid Lq1 adheres to the camera 70 or the vaporized component of the treatment liquid Lq1 adheres to the camera 70. Therefore, even if the treatment liquid Lq1 contains hydrofluoric acid, the camera 70 is unlikely to be corroded.

  The camera 70 may be fixed substantially immovable in the processing unit 1B or may be fixed movably.

  Further, the mirror 75 does not necessarily have to be fixed to the fixing member 62 of the processing liquid supply unit 60, and may be fixed to the fixing member 32 of the processing liquid supply unit 30 like the camera 70 of the processing unit 1A. According to this, since the mirror 75 can be brought closer to the nozzle arm 321, the imaging direction from the imaging position can be more easily aligned with the circumferential direction.

<Machine learning>
In the above-described example, the control unit 9 performs the image processing on the captured image IM1 to determine the presence / absence of ejection of the treatment liquid Lq1 and the presence / absence of liquid splash. However, the control unit 9 may make the determination using machine learning.

  FIG. 16 is a diagram schematically showing an example of the internal configuration of the control unit 9. The control unit 9 includes a classifier 91 and a machine learning unit 92. The captured images IM1 from the camera 70 are sequentially input to the classifier 91. The classifier 91 classifies each input captured image IM1 into a category related to the ejection state of the ejection nozzle 31. Categories can also be called classes. As the categories, a first category indicating discharge stop, a second category indicating normal discharge, and a third category indicating liquid splash can be adopted.

  The classifier 91 is generated by the machine learning unit 92 using a plurality of teacher data. That is, it can be said that the classifier 91 is a machine-learned classifier. The machine learning unit 92 uses, for example, a neighborhood method, a support vector machine, a random forest or a neural network (including deep learning) as a machine learning algorithm. Since the neural network automatically generates the feature quantity, the designer does not need to determine the feature vector.

  The teacher data includes image data and a label indicating to which category the image data is classified. The image data is a captured image captured by the camera 70 and is generated in advance. A correct category is given as a label to each image data. This giving can be performed by the operator. The machine learning unit 92 performs machine learning based on these teacher data to generate the classifier 91.

  As an example, a classifier 91 that classifies frames by the neighborhood method will be described. The classifier 91 includes a feature vector extraction unit 911, a determination unit 912, and a storage medium in which the determination database 913 is stored. Each frame of the captured image from the camera 70 is sequentially input to the feature vector extraction unit 911. The feature vector extraction unit 911 extracts the feature vector of the captured image IM1 according to a predetermined algorithm. This feature vector is a vector indicating a feature amount according to the ejection state of the ejection nozzle 31. A known algorithm can be adopted as the algorithm. The feature vector extraction unit 911 outputs the feature vector to the determination unit 912.

  The determination database 913 stores a plurality of feature vectors (hereinafter referred to as reference vectors) generated from a plurality of teacher data by the machine learning unit 92, and the reference vectors are classified into each category. Specifically, the machine learning unit 92 applies the same algorithm as the feature vector extraction unit 911 to a plurality of teacher data to generate a plurality of reference vectors. Then, the machine learning unit 92 gives the label (correct category) of the teacher data to the reference vector.

  The determination unit 912 classifies the captured image IM1 based on the feature vector input from the feature vector extraction unit 911 and the plurality of reference vectors stored in the determination database 913. For example, the determination unit 912 may specify the reference vector having the closest feature vector, and classify the captured image IM1 into the specified reference vector category (nearest neighbor method). Thereby, the determination unit 912 can classify the captured image input to the classifier 91 (feature vector extraction unit 911) into categories.

  The control unit 9 classifies each captured image IM1 into one of the first category to the third category by the classifier 91. This classification means that it is determined whether or not the ejection state amount of the treatment liquid Lq1 is within an appropriate range. Since the classification is performed by machine learning, the abnormality can be detected with high accuracy.

<Input to classifier>
Although the entire area of the captured image IM1 is adopted as the input data to the classifier 91 in the above example, the input data is not limited to this. For example, the control unit 9 may cut out an image of the ejection determination region R2 from the captured image IM1 and input the image to the classifier 91. In this case, the image showing the ejection determination region R2 is also used as the learning data input to the machine learning unit 92.

  According to this, the classifier 91 can perform the classification by removing the influence of the region that is less relevant to the ejection state, and thus the classification accuracy can be improved.

  If the ejection determination region R2 has a width corresponding to two or more pixels in the vertical direction, the pixels arranged in a line in the vertical direction of the ejection determination region R2 are input data to the classifier 91. You may employ | adopt the integral value group which contains the integral value which is the sum total of a value for every column.

<Server>
In the above example, the control unit 9 provided in the substrate processing apparatus 100 generates the classifier 91 by machine learning, and the classifier 91 classifies the frames. However, at least a part of the machine learning function (the classifier 91 and the machine learning unit 92) by the control unit 9 may be provided in the server.

  Although the embodiment of the present substrate processing apparatus has been described above, various modifications other than those described above can be made to this embodiment without departing from the spirit thereof. The various embodiments and modified examples described above can be implemented in an appropriate combination.

  Although the semiconductor substrate is used as the substrate W in the description, it is not limited to this. For example, a substrate such as a photomask glass substrate, a liquid crystal display glass substrate, a plasma display glass substrate, an FED (Field Emission Display) substrate, an optical disk substrate, a magnetic disk substrate, or a magneto-optical disk substrate may be adopted. Good.

20 substrate holding part 31 first nozzle (ejection nozzle)
33, 63 moving mechanism 40 cup member (processing cup)
44 Lifting mechanism 66, 69 Second nozzle (discharge nozzle)
70 Camera 75 Mirror 91 Notification Unit 91 Classifier 100 Substrate Processing Device W Substrate

Claims (12)

A holding step of holding the substrate on the substrate holding part,
A rotating step of rotating the substrate by rotating the substrate holder,
A raising step of raising a cup member surrounding the outer periphery of the substrate holding portion to position an upper end of the cup member at an upper end position higher than an upper surface of the substrate held by the substrate holding portion;
A bevel processing step of discharging a processing liquid from the discharge port of the nozzle located at a position lower than the upper end position to the end portion of the upper surface of the substrate held by the substrate holding unit;
A captured image obtained by allowing the camera to capture an imaged region including the processing liquid ejected from the ejection port of the nozzle, the imaged region being viewed from an image capturing position above the substrate held by the substrate holder. An imaging step for acquiring
And a monitoring step of determining a discharge state of the processing liquid based on the captured image. The substrate processing method according to claim 1, wherein
The monitoring step includes
It is determined whether or not the first statistic of the luminance value of the pixel in the ejection determination region, which is located immediately below the nozzle in the captured image and is longer in the horizontal direction than in the vertical direction, is greater than or equal to a first threshold, A substrate processing method comprising a step of determining that the processing liquid is ejected from the nozzle when the first statistic is equal to or more than the first threshold value. The substrate processing method according to claim 2, wherein
The monitoring step includes
It is determined whether the first statistic is equal to or more than a second threshold larger than the first threshold, and when the first statistic is equal to or more than the second threshold, the treatment liquid is the substrate. A method for processing a substrate, comprising a step of determining that liquid splashing on the upper surface is generated. The substrate processing method according to claim 2 or 3, wherein
The monitoring step includes
Based on the brightness value of the pixel in the liquid splash determination region set on the downstream side in the rotation direction of the substrate with respect to the ejection determination region in the captured image, the treatment liquid is applied on the upper surface of the substrate. A substrate processing method, comprising the step of determining whether or not splashing liquid splashes have occurred. The substrate processing method according to claim 4, wherein
The substrate processing method, wherein the liquid splash determination region is not set on the upstream side in the rotation direction of the substrate with respect to the ejection determination region in the captured image. The substrate processing method according to any one of claims 2 to 5,
The substrate processing method, wherein the ejection determination region is set at a position including a part of the processing liquid reflected on the upper surface of the substrate by specular reflection in the captured image. The substrate processing method according to claim 6, wherein
The substrate processing method, wherein the exposure time of the camera is set to be equal to or longer than a time required for the substrate to rotate once. The substrate processing method according to claim 6, wherein
Regarding the luminance value of the pixel in the ejection determination region in the captured image obtained by integrating or averaging a plurality of captured images acquired by the camera within a time period equal to or longer than the time required for the substrate to rotate once, A substrate processing method for obtaining the first statistic. The substrate processing method according to claim 1, wherein
In the monitoring step,
A substrate processing method, wherein the captured image is classified into a category indicating a discharge state of a processing liquid discharged from the tip of the nozzle by a classifier that has been machine-learned. The substrate processing method according to claim 9,
In the monitoring step,
A substrate processing method, wherein an ejection determination region located directly below the nozzle and longer in the horizontal direction than in the vertical direction is cut out from the captured image, and the image of the cut out ejection determination region is input to the classifier. The substrate processing method according to any one of claims 1 to 10, wherein
The captured image includes a part of the periphery of the substrate,
The bevel processing step,
Determining a part of the peripheral edge of the substrate based on the captured image;
A substrate processing method, comprising a step of moving the nozzle from a peripheral edge position of the substrate to a processing position at a central portion of the substrate by a predetermined width. A substrate holding unit for holding the substrate and rotating the substrate;
A cup member that surrounds the outer periphery of the substrate holder,
An elevating mechanism that raises the cup member to position the upper end of the cup member at an upper end position higher than the upper surface of the substrate held by the substrate holding portion;
A nozzle that has a discharge port at a position lower than the upper end position, and discharges a processing liquid from the discharge port to an end of the upper surface of the substrate held by the substrate holding unit;
A captured image is obtained by capturing an imaged region including the processing liquid ejected from the ejection port of the nozzle, the imaged region being viewed from an image capturing position above the substrate held by the substrate holder. A camera,
An image processing unit that determines an ejection state of the processing liquid based on the captured image.

Images